딥러닝논문 (2) 썸네일형 리스트형 [Deep learning 논문 읽기] 리뷰할 논문 목록 (주1 예정) stylegan2 단점 보완 논문 : 너무 많은 데이터가 필요하단 점을 보완한 논문 1. Training Generative Adversarial Networks with Limited Data arxiv.org/pdf/2006.06676.pdf Metric 관련 2. generator metric : GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium (Learned Perceptual Image Patch Similarity (LPIPS)) arxiv.org/abs/1706.08500 GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash.. [Deep learning 논문 읽기] StyleGAN2 Abstract 뜯어보기 We expose and analyze several of its characteristic artifacts, and propose changes in both model architecture and training methods to address them stylegan2에서는 지난 버전에서 몇 가지 특징적인 결함들을 분석했고, 이에 대한 방법으로 모델 구조와, 훈련 방법에 있어 변화를 제안한다. In particular, we redesign the generator normalization, revisit progressive growing, and regularize the generator to encourage good conditioning in the mapp.. 이전 1 다음
